ICEinLAVA's Domain

Somehow, by virtue of attrition I guess, I seem to have graduated to the position of an elder audio statesman. Suddenly I find myself the butt of “doing sound at the Last Supper” jokes and being pumped by young roadies about what it was like “in the day.”

This is not necessarily a bad thing, as I am still doing what I love more than a quarter of a century after my first gig, despite Mom’s weekly chiding about when I’m going to get a real job.

I am forever baffled, though, when young engineers come to me for tips or advice, and I am greeted with a deer in the headlights stare when I tell them that the most important button on the console, at least in my opinion, is the phase button. I’m usually met by, “I always wondered what that button does.”

Riding the Wave

Most of us immediately hear left/right program out of phase, that nauseating sensation that something in your head’s not right. Yet two or more channels out of phase in a relatively coherent mix can be much harder to hear or sense.

The true nature and efficiency of phase was best demonstrated to me in an L.A. Times article days after the 1994 Northridge earthquake. In the article, it explained that there was an almost simultaneous earthquake in Hollywood, with the two low frequency waves running towards each other through the soft sandy soil of the San Fernando Valley. This did a great deal to explain why there was massive destruction in one area, while mere blocks away, little was touched. The summing of the two waves amplified the quake in some areas while the cancellation nullified the seismic event in others. As this is one of the most basic laws of physics in audio, I can’t think of a better, larger or easier example to grasp.

“In the day,” when I started in the studio in the mid-’70s, we were still working with vinyl, and the X/Y display was god. The reason for this was that the cutting heads on the lathes moved in four directions. Left, right for stereo information and up, down for amplitude. Out of phase 180 sent the lathe head straight up out of the acetate. When your mastering engineer informed you of this, the only option was to go back and re-mix it, impacting the budget and release date, with your name maligned in memos between the producer and record executives for awhile, if not the length of your career. (More than one career was definitely shortened this way, and many more after the introduction of the Aphex Aural Exciter, a great tool that could be very mastering unfriendly.) So you just made sure your mix was phase coherent.

With the advent of CDs, phase has unfortunately become somewhat superfluous, as digital just doesn’t care. It can and will reproduce whatever you put into it. As a result, a good deal of information on many of today’s CDs, when listened to in mono, simply disappears.

Now, if the recording industry isn’t that concerned about it anymore, why should we in the touring industry care? Well, we’re still dealing with large three-dimensional environments, multiple point sources, crossover points and maximizing component amplification. What does that mean? (At this point, a caveat to Yamaha users: one of the few ergonomic flaws on their consoles is the placement of the phase button next to the 30db pad. Be mindful not to disengage the pad when using the phase button!)

For starters, let’s say you have a 40-box hang with two 15s in each box (80) and 20 subs on the floor with two 18s each (40). Phase is the electronic command to push or pull the cone of the speaker. If your kick drum is pushing and your bass is pulling, the speakers are canceling the two exactly where you’re looking for punch—120 speakers putting out next to nothing with your amps taxed to the limit. Trust me, you can jerk the EQs around all night and not much is going to change except the temperatures of your amps. When you take into consideration some of the new intelligent crossover and amp systems with steering subs, etc., many of their functions are accomplished with phase manipulation, and a less coherent mix can confuse and undermine the system. Hopefully, I just heard a couple of light bulbs come on. As simple as it seems, we don’t always think of this stuff. So where else can we improve phase coherency?

Mono inputs from sources, i.e. mic and DI, will still read if out of phase, but never get in your face. (As phase is most apparent in the lower spectrum, let “fatter in phase and thinner out” be your guide). Monitor guys working with ears, be careful, as finding one of these can be dramatic to the tune of 3db or twice as loud, especially if you’ve been pushing it. It may not seem terribly important for hard left, right channels such as keys, guitars, loops etc. to be phase coherent in the house, with 40 feet of separation between your stacks, until you consider lobing. Lobing occurs when your stacks or components cross and either sum or cancel. Out of phase information will exacerbate this phenomenon, making the lobing more extreme. Also, those local TV crews, the ones who show up four minutes before house lights and want a “mono” feed, will expose any phase issues on the 11 o’clock news. (As to loops in particular, these are usually generated in the recording studio by a second engineer in the wee hours of the morning after a long mix session. I would suggest checking the phase of each one, as I have found discrepancies from one loop to another on projects I’ve been involved with.)

Overheads, by the nature of the term, are not cymbal mics. Everyone realizes that you get more low end out of a snare if you flip the bottom mic out of phase to the top. The same applies to taking the time to make sure your snare top and tom mics are in phase with your overheads. Run your high pass lower and you’ll find more air around the drums and tight gating less apparent. Think Zeppelin!

Turning It Around

So what about phase problems the 180 button just can’t cure? For that, we resort to time alignment, using delay for phase correction of less than or greater than 180 degrees from multiple sources at different distances. Two good examples of this are delaying the house P.A. to the backline and delaying the lead singer’s wedges to the sidefills. In this manner, all the information arrives at the same place at the same time, in phase, and the delay is negligible.

So everything has to be in the same phase all the time, right? Not necessarily. Within the mix itself, you want anything that can sum to do so. However, there are instances where phase cancellation can be tremendously helpful. Let’s look at a few examples of this.

1. Hot mics on the downstage edge can sometimes be hard to hold in the P.A., and even harder with a loud stage mix. Throwing that channel out of phase in the house, in theory, can sometimes buy you 3db more gain before feedback. This doesn’t always work, but more times than not it helps.

2. Here’s a good monitor trick Dave Shadoan and I tripped over on a Babyface tour a few years back. We had six background singers who took leads throughout the show in a straight line with their wedges spaced about six inches apart. All the mixes had similar information and were all pretty hot. Dave had rung them all out and they seemed stable until we put all six of them up there at once. It drove us nuts, until as a last ditch effort, we flipped the phase of every other vocal on the monitor console. The gain before feedback ratio improved to the point that they were asking to be turned down.

3. Experiment with the phase between your subs and lows. It’s in the nature of EQ (which is really what a crossover is) to create a reciprocal curve or bump on either side of your cut or crossover point. This can sometimes cause unwanted summing in strange places between the components that switching the phase may alleviate, giving you more punch. For whatever reason, I’ve found this trick to be the most environment-dependant, and would be different from gig to gig.

So there you go. My pet peeve purged. The button with the Greek symbol on it need not be Greek to you. Put on your cans, pan to center, flip the phase and check it out.